Narrow Results

A Pharmaceutical formulation is composed of several formulation factors and process variables. Quantitative model based pharmaceutical formulation involves establishing mathematical relations between the formulation variables and the resulting responses, and optimizing the formulation conditions. In a formulation system involving several objectives, the desirable formulation conditions for one property may not always be desirable for other characteristics, thus leading to the problem of conflicting objectives. Therefore, efficient modeling and optimization techniques are needed to devise an optimal formulation system. In this work, a novel method based on radial basis function network (RBFN) is proposed for modeling and optimization of pharmaceutical formulations involving several objectives. This method has the advantage that it automatically configures the RBFN using a hierarchically self organizing learning algorithm while establishing the network parameters. This method is evaluated by using a trapidil formulation system as a test bed and compared with that of a response surface method (RSM) based on multiple regression. The simulation results demonstrate the better performance of the proposed RBFN method for modeling and optimization of pharmaceutical formulations over the regression based RSM technique.

In China, the evolution of inter-provincial migration patterns is from policy orientation to individual selection and hotspot attraction during 1950–2010. These years, China’s abundant regional labor flow brings prosperity to several famous high-growth regions, but also causes the problem of unbalanced development. It needs systematic analysis to do effective guidance and management of population mobility. This paper tries to analyze the driving and dragging force in China’s inter-provincial migration flows with improved multilateral migration model stemming from individual behavior of potential migrants. It revealed the problem of imbalanced regional labor flow has been exacerbated in China since the 1990s, and the driving force for typical hotspots is different, as high income force for Beijing, Shanghai, and low barrier for Guangzhou, Zhejiang. The results of analyzing indicates the the distinct regional mobility barriers help to alleviate the problem of labor loss in some provinces since 1980s, but the regional imbalance in population migration continues to grow in recent years. Finally, with regression, we reveal the essential element affecting regional mobility barriers, and it could explain 70.8% of the influence factors. It also suggests that, for the provinces with labor loss, improving economic, welfare environment will mitigate the problem effectively.

In the presence of a number of algorithms for classification and prediction in software engineering, there is a need to have a systematic way of assessing their performances. The performance assessment is typically done by some form of partitioning or resampling of the original data to alleviate biased estimation. For predictive and classification studies in software engineering, there is a lack of a definitive advice on the most appropriate resampling method to use. This is seen as one of the contributing factors for not being able to draw general conclusions on what modeling technique or set of predictor variables are the most appropriate. Furthermore, the use of a variety of resampling methods make it impossible to perform any formal meta-analysis of the primary study results. Therefore, it is desirable to examine the influence of various resampling methods and to quantify possible differences. Objective and method: This study empirically compares five common resampling methods (hold-out validation, repeated random sub-sampling, 10-fold cross-validation, leave-one-out cross-validation and non-parametric bootstrapping) using 8 publicly available data sets with genetic programming (GP) and multiple linear regression (MLR) as software quality classification approaches. Location of (PF, PD) pairs in the ROC (receiver operating characteristics) space and area under an ROC curve (AUC) are used as accuracy indicators. Results: The results show that in terms of the location of (PF, PD) pairs in the ROC space, bootstrapping results are in the preferred region for 3 of the 8 data sets for GP and for 4 of the 8 data sets for MLR. Based on the AUC measure, there are no significant differences between the different resampling methods using GP and MLR. Conclusion: There can be certain data set properties responsible for insignificant differences between the resampling methods based on AUC. These include imbalanced data sets, insignificant predictor variables and high-dimensional data sets. With the current selection of data sets and classification techniques, bootstrapping is a preferred method based on the location of (PF, PD) pair data in the ROC space. Hold-out validation is not a good choice for comparatively smaller data sets, where leave-one-out cross-validation (LOOCV) performs better. For comparatively larger data sets, 10-fold cross-validation performs better than LOOCV.

Wildlife managers need to evaluate the regional risk of damage by big game in any cultivated plot. Nevertheless, such an evaluation can be biased by nonlinearity, a common difficulty when facing ecological problems. We propose a model for the impact of wild boars on cultivated fields, based on artificial neural networks, with a backpropagation algorithm. The first model, predicting the frequency of impact on a particular plot, gives a good determination coefficient (R²=0.91). The second model, predicting the presence or absence of impact during a particular week, gives over 80 % correct results.

Lateral distortional buckling (LDB) mode has been investigated as one of the failure modes in steel I-beams. The LDB mode in such beams is known by simultaneous lateral deflection, twist, and cross-sectional change due to web distortion. However, many studies have been conducted to investigate the behavior of this failure mode; so far, no formula has been found to calculate the ultimate LDB resistance of I-beams, which also takes into account the effect of different loading conditions. Consequently, in the current paper, by conducting an extensive parametric study, it was tried to investigate the effect of all main parameters as well as the effect of different loading conditions on the ultimate LDB resistance of I-shaped beams. Then, based on the provided database, the artificial neural network (ANN) method was employed, and based on it, a high-precision formulation was proposed to predict the ultimate LDB strength of steel I-beams. In addition to the ANN method, a regression-based formula was also developed as a classical method to examine the differences between the two methods. Finally, the proposed formulas were compared with other existing formulas for estimating the LDB strength. The results showed that the proposed formula based on ANN not only presents a reasonable accuracy compared to the existing formulations but also can be used by engineers as practical equations in the design of I-beams.

Drilling procedures are important to optimize and ensure the strongest fixation in bone fracture treatment and reconstruction surgery. The mechanistic force models currently available for bovine bones, human spines and human mandibles are not relevant to perform drilling through human femurs. The present study addresses this lack of information and aims to develop the predictive force models for drilling human femurs at different regions and directions. In this study, 10 freshly harvested cadaveric human femurs were included, and a surgical drill bit of 3.2mm diameter was used to make 4mm deep holes. Different spindle speeds (500, 1000 and 1500rpm), feed rates (40, 60 and 80mm/min), and apparent density between 0.98 and 1.98g/cm³ were considered. The optimal parameters $f_3{s}_3$, $f_2{s}_3$, and $f_1{s}_3$ respectively obtained for longitudinal, radial, and circumferential direction could minimize the thrust forces in bone drilling by up to 7.70, 10.50, and 16.20 N, respectively. Validation study demonstrated that the force model developed could predict the thrust force from computed tomography data sets of the patient, only with 5.05%, 6.74%, and 4.91% as a maximum error in longitudinal, radial, and circumferential direction. This important tool can assist to perform complicated surgical operations.

We present a method for gene network inference and revision based on time-series data. Gene networks are modeled using linear differential equations and a generalized stepwise multiple linear regression procedure is used to recover the interaction coefficients. Our system is designed for the recovery of gene interactions concurrently in many gene regulatory networks related by a tree or a more general graph. We show how this comparative framework can facilitate the recovery of the networks and improve the quality of the solutions inferred.

This paper reports on the development of a computer model that will predict both overall project and activity duration, based on a number of pre-determined project characteristics. Fifty-six programmes of work were obtained. The data from the programmes of work of fifty of these buildings, encompassing a total of 11 different project types, were analysed, and used to develop the proposed model. Multiple linear regression analysis of the data showed that the duration and time lags of between 20 (for a single storey building) and 39 (for a seven-storey building) standardised activity groups, can be predicted using combinations of the twenty one most influential project variables.

The regression equations produced were tested on all of the activity groups for six new projects to determine their accuracy. The absolute percentage error in predicting overall duration varied between 0.38% and 6.68%. The mean absolute error in predicting the duration of activity groups varied between 1.38% and 22%. The accuracy in predicting overall duration was comparable with limited information available from previous studies, but the high level of detail in the programme generated means that the model is more flexible and capable of a broader range of applications than previous models.

Multiple regression is not reliable to recover predictor slopes within homogeneous subgroups from heterogeneous samples. In contrast to Monte Carlo analysis, which assigns completely to the first-specified predictor the variation it shares with the remaining predictors, multiple regression does not assign this shared variation to any predictor, and it is sequestered in the residual term. This unassigned and confounding variation may correlate with specified predictors, lead to heteroscedasticity, and distort multicollinearity. I develop and test an iterative, sequential algorithm to estimate a two-part series of weighted least-square (WLS) multiple regressions for recovering the Monte Carlo predictor slopes in three homogeneous subgroups (each generated with 500 observations) of a heterogeneous sample $(n=1,500)$. Each variable has a different nonnormal distribution. The algorithm mines each subgroup and then adjusts bias within it from 1) heteroscedasticity related to one, some, or all specified predictors and 2) “nonessential” multicollinearity. It recovers all three specified predictor slopes across the three subgroups in two scenarios, with one influenced also by two unspecified predictors. The algorithm extends adaptive analysis to discover and appraise patterns in field research and machine learning when predictors are inter-correlated, and even unspecified, in order to reveal unbiased outcome clusters in heterogeneous and homogeneous samples with nonnormal outcome and predictors.